Method

• Conducted Feature matrix of the four AI Chatbots 

• Created general scenarios and UX tasks that each team member tried in each chatbot 

• Prompting Techniques used: 

• Role Assignment 

• Instruction

• Examples

• Context

• Priming

Competitive Analysis of AI Chatbots (2023)

Purpose: To evaluate various AI softwares currently on market to establish pros and cons of each.  

Feature Matrix Highlights: Many of the features overlap between AI softwares so ultimate decision making on use depends on purpose and prompt. 

Recommendations: Overall, Bard* [now Gemini] and ChatGPT come out on top as safe choices for UX researchers and general consumers to use. 

Results & Recommendations

In general for the average consumer:

• More widespread usage of Bard-particularly in a source collection processes

• ChatGPT for more technical knowledge and coding

• Bing Chat for more informal/conversational usage

• Claude 2- Potential future use after updates

For the UX Researcher:

• Conduct further research on the use of priming, context, instructions, and examples to improve the performance of AI chatbots

• Develop and test new methods for evaluating the performance of AI chatbots in real-world settings

• Collaborate with industry partners to develop new AI chatbot applications that can benefit users in more industry specific contexts